Abstract：Image dehazing is researched based on the mean clustering and loop guided filtering(MCLGF). Firstly, image dehazing optical model was established. Secondly, two sub feature vectors were selected randomly as initial cluster center, image was divided into the sky and non sky area from the classification results based on mean clustering, atmospheric light value was obtained from the sky area pixel. Thirdly, loop guided filtering was achieved the details of smoothing and edge orientation, and difference output and input image of guided filtering was minimize with minimization cost function. Finally, the algorithm process was given. Simulation results show mean clustering and loop guided filtering could reduce the haze of image effectively, and structural similarity is 0.98, information fidelity is 0.96, information entropy is 9.12 and regular gradient mean is 0.85, the four evaluation indexes are better than the other algorithms.
［1］程磊, 刘勇军. 基于改进暗通道算法的图像去雾研究［J］. 计量学报, 2019, 40(2): 220-224.
Cheng L, Liu Y Jun. Image Dehazing Based on Improved Dark Channel Prior Algorithm［J］. Acta Metrologica Sinica, 2019, 40(2): 220-224.
［2］Fattal R. Single image dehazing［J］. ACM Transactions on Graphics, 2008, 27(3): 1-9.
［3］Tan R T. Visibility in bad weather from a single image［C］//Proceedings of the 2008 IEEE Conference on Computer Vision and Pattern Recognition, Anchorage, USA, Jun 24-26, 2008. Piscataway, NJ, USA: IEEE, 2008: 1-8.
［4］Tarel J P, Hautiere N. Fast visibility restoration from asingle color or gray level image［C］//Proceedings IEEE-Conference on Computer Vision. Kyoto, Japan, IEEE, 2009: 1701-1708.
［5］He K M, Sun J, Tang X O. Single image haze removal using dark channel prior［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2011, 33(12): 2341-2353.
［6］Nishino K, Kratz L, Lombardi S. Bayesian defogging［J］. International Journal of Computer Vision, 2012, 98(3): 263-278.
［7］Laurence M L, Sabine S. High dynamic range image rendering with a retinex-based adaptive filter［J］. IEEE Transactions on Image Processing, 2006, 15(9): 2820-2830.
［8］胡平, 杨旭东. 高速公路能见度快速检测算法［J］. 公路交通科技, 2017, 34(4): 115-122.
Hu P, Yang X D. An Algorithm for Fast Detecting Expressway Visibility［J］. Journal of Highway and Transportation Research and Denelopment, 2017, 34(4): 115-122.
［9］陈勇, 郝裕斌, 张开碧. 结合K均值聚类的暗通道去雾改进算法［J］. 半导体光电, 2016, 37(6): 876-881.
Chen Y, Hao Y B, Zhang K B. Improvement of Haze Removal Method Combining Dark Channel Priority with K-means Clustering Algorithm［J］. Semiconductor Optoel-ectronics, 2016, 37(6): 876-881.
［10］陈书贞, 任占广, 练秋生. 基于改进暗通道和导向滤波的单幅图像去雾算法［J］. 自动化学报, 2016, 42(3): 455-465.
Chen S Z, Ren Z G, Lian Q S. Single Image Dehazing Algorithm Based on Improved Dark Channel Prior and Guided Filter［J］. Acta Automatica Sinica, 2016, 42(3): 455-465.
［11］韩正汀, 路文, 杨舒羽, 等. 基于导向滤波优化的自然图像去雾新方法［J］. 计算机科学与探索, 2015, 9(10): 1256-1262.
Han Z T, Lu W, Yang S Y, et al. Improved natural image dehazing algorithm based on guided filtering［J］. Journal of Frontiers of Computer Science & Technology, 2015, 9(10): 1256-1262.
［12］胡子昂, 王卫星, 陆健强, 等. 视觉信息损失先验的图像分层去雾方法［J］. 中国图象图形学报, 2016, 21(6): 711-722.
Hu Z A, Wang W X, Lu J Q, et al. Image dehazing using visual information loss prior［J］. Journal of Image and Graphics, 2016, 21(6): 711-722.
［13］姚成乾. 基于最大熵区间分析的测量不确定度评定［J］. 计量学报, 2019, 40(1): 172-176.
Yao C Q. Measurement Uncertainty Evaluation Based on Maximum Entropy Interval Analysis［J］. Acta Metrologica Sinica, 2019, 40(1): 172-176.
［14］田源. 基于混合远景和近景区域暗通道算法的图像去雾研究［J］. 计量学报, 2019, 40(4): 583-588.
Tian Y. Image Dehazing Based on Near and Far Scene Dark Channel Prior［J］. Acta Metrologica Sinica, 2019, 40(4): 583-588.
［15］张淑清, 乔永静, 姜安琦, 等. 基于CEEMD和GG聚类的电能质量扰动识别［J］. 计量学报, 2019, 40(1): 49-57.
Zhang S Q, Qiao Y J, Jiang A Q, et al. Power Quality Disturbance Identification Based on CEEMD and GG Clustering［J］. Acta Metrologica Sinica, 2019, 40(1): 49-57.
2021, Vol. 42 
